Radiation from the big bang

The idea of another fossil, a remnant thermal sea of radiation that might be expected to accompany the production of helium in a hot big bang, was clearly expressed in the literature, including Alpher and Herman (1948, 1950).41

Osterbrock recalls (p. 88) hearing Gamow lecture at the University of Michigan in the summer of 1953. In the published version of these lectures, Gamow (1953a) described his ideas about element formation in a hot big bang. He presented his 1948 argument (based on the condition in footnote 10 on page 30) that, for production of a reasonable amount of deuterium that would mostly burn to helium, "about the right density" of baryons when the

41 Gamow (1948a,b) expressed the same idea, but less directly. He pointed out that the radiation would be present after element formation, and that it could be an important dynamical actor in structure formation. This is discussed in Section 3.6.4. Gamow continued to emphasize the potential importance of the radiation, but his arguments were unfortunately confused by the error noted in the next footnote.

temperature was 109 K is 2 x 1018 baryons cm-3 (calculated from the numbers he gives). Under these conditions the temperature would have dropped to T0 = 7K at Gamow's estimate of the present baryon density (equation 3.6). This is close to what Alpher and Herman had obtained earlier from Gamow's condition (equation 3.7). Gamow did not write down this last step, however. Earlier in the lectures he discussed the idea of a sea of thermal radiation that cools as the universe expands, and he considered the effect of the mass equivalent of the radiation on the rate of expansion. He argued from a consideration of expansion time that when the mean mass density had dropped to 10-25 g cm-3 the radiation temperature would have dropped to 320K. That extrapolates to present value T0 = 7K at his estimate of the present density. This is close to the Alpher and Herman value, and to what was later measured, but the calculation is unfortunately wrong.42 The CMBR idea nevertheless is in these lectures, though we have not encountered anyone who noticed until much later.

An illustration of the low visibility of the CMBR idea is the absence of any mention of the idea of radiation from the early universe in Bondi (1952, 1960a). Another useful indication of what people were thinking comes from the proceedings, or published records, of international conferences. Most exchanges of ideas at these events tend to be in informal discussions, and the formal lectures are not always close to what appears in the proceedings, but the published versions do show what people considered worth recording. The proceedings of the Solvay conference, La Structure et l'Evolution de l'Univers (the eleventh in a distinguished series of meetings on major advances in physics), in Brussels in 1958; the 9th International Astronomical Union Symposium, Paris Symposium on Radio Astronomy (Bracewell 1959), in Paris in 1958; and the 15th International Astronomical Union Symposium, Problems of Extra-Galactic Research (McVittie 1962), in Santa Barbara, California in 1961, all include papers on issues in cosmology as well as on advances in the astronomy of radio sources, stars, and galaxies. In these volumes we find no mention of the idea that space might be filled with a sea of microwave radiation, perhaps one left from the early universe.

There are accounts of informal discussions of the idea. Alpher and Herman (2001) recall asking radio astronomers about the possibility of detecting the radiation. Tayler (1990), in his recollections of the work with Hoyle in 1964 on helium production, mentions their thoughts - which did not enter

42 The calculation assumed space curvature in the expansion equation (G.1) becomes the dominant term just when the mass densities in matter and radiation are equal. That need not be so. Gamow (1953b, 1956) repeated this calculation and derived from it the present temperature, To ~ 7K. But, as opposed to his now well-tested argument for light element formation, this calculation is not of lasting interest.

their paper - about remnant radiation from a big bang. Hoyle (1981) recalls discussions of the possible temperature of a sea of microwave radiation in conversations with Gamow and Dicke. In a very readable popular book, The Creation of the Universe, Gamow (1952) describes the cooling of a sea of thermal radiation in an expanding universe. He notes, as an example, that a universe with mass dominated by radiation cools to a temperature of 50 K at three thousand million years after the big bang, a common estimate then for the age of the universe. (The method of calculation is indicated in footnote 1 on page 26.) The calculation is not directly relevant, because the mass of our universe is not now dominated by radiation, but the pointer to a sea of thermal radiation is quite direct.

On the issue of detectability we have an account by Virginia Trimble of Gamow's encounter with Joe Weber, who had expertise in microwave technology.

Joe Weber was an amateur radio operator in his early teens and, at the time of the Sicilian invasion, was the skipper of one of the first submarine chasers to have a 6-cm radar (SC 690). As the war wound down, the Navy moved him to a desk job in Washington in electronic countermeasures, largely to descope the effort, but also to hand out some grants. When he decided to resign his commission (as lieutenant commander), several grantee organizations offered him jobs, but he accepted instead a full professorship of electrical engineering at the University of Maryland. The fall 1948 appointment was contingent on his obtaining a PhD in something quite soon, since his highest degree was a 1940 BS from the US Naval Academy.

Thus summer 1949 found Weber visiting Washington-area universities in search of a PhD project and advisor. One of the first places he visited was George Washington University, and one of the people he talked with there was George Gamow. "Do you have any interesting thesis problems?" Weber enquired. "What can you do, young man?" responded GG. "I'm a microwave spectroscopist," said JW. "No, I don't think of any interesting problems" concluded Gamow. So Weber went on to Catholic University, where he completed a 1950 PhD dissertation (Weber 1951) with Keith Laidler on the inversion spectra of normal and deuterated ammonia. Since Weber at the time knew about the technology for detecting faint radio signals, whether the story is funny depends on whether you think Gamow should have had radiation from the early universe in mind in 1948. It is, of course, a secondhand story, but I was married to Joe from 1972 until his death in the year 2000, and men, as you probably know, like to tell war stories. There is also a good one about the inhabitants of Tonga Tabu, following the sinking of the Lexington in the battle of the Coral Sea in May 1942.

For other aspects of this issue see Trimble (2006) and Burke's account on page 181.

Our conclusion is that the idea that space might be filled with thermal radiation left from the early stages of expansion of the universe - what we now term the CMBR - was "in the air" in the early 1960s. But it was less visible than other issues in cosmology, particularly the debates on the relative merits of the big bang and steady state scenarios.

The first ground-based measurements that, with the full benefit of hindsight, might be said to have offered a suggestive indication of detection of this microwave radiation were the Bell Laboratories Echo and Telstar communications experiments we discussed in Section 3.5. They are described in more detail in the next chapter.

There were earlier measurements that placed what were later recognized to be interesting upper bounds on the CMBR temperature. We have mentioned the Dicke et al. (1946) limit, T0 < 20 K at wavelengths of 1-1.5 cm. They referred to "radiation from cosmic matter," not a cosmological model. Dicke (1946a) also used his radiometer to measure the radiation from the Moon and the Sun at 1.25 cm.

Covington (1950), in Canada, was studying bursts of radio radiation from the Sun at longer wavelength and the correlation of these bursts with radiation produced by disturbances of Earth's upper atmosphere (the ionosphere). That required an estimate of radiation from Earth's undisturbed atmosphere and beyond. Covington reported that at 10.7-cm wavelength this radiation is small, "not more than about 50° K." Within the experimental error, ±25 K, this is not significantly different from zero, or from the temperature, 50 K, in the example cosmological model in Gamow's (1952) popular book. We have seen no notice taken of the latter coincidence.

Haruo Tanaka et al. (1951), at Nagoya University in Japan, using a square horn antenna at 8-cm wavelength, obtained a better constraint, that the incident radiation at zenith (including what is produced by the atmosphere) is no more than about 5K. Tanaka (1979) offers this comment on their measurement.43

14 years before [Penzias and Wilson's discovery], we measured the temperature of sky at the wavelength of 8cm, and estimated it to be between 0 and 5 K. ... The measurement [of the sky temperature at zenith] was made for an absolute calibration of the intensity of solar radio waves. ... Except for a parabolic reflector requiring accurate shaping, our instruments were all handmade: we obtained the necessary parts from the disposal goods of the army. ... Since we could not calculate a gain of the parabolic antenna, we built a pyramidal horn antenna whose length was 2 m. ... At that time A. E. Covington in Ottawa, Canada, had been observing the solar radio waves at 10.7 cm since 1947. He calibrated the solar flux using the sky temperature of 50 K. However it seemed too high for us, and we decided to measure the sky temperature by ourselves. ... I understand that 0-5K and 3.5 ± 1K are

43 We are grateful to Eiichiro Komatsu and Tsuneaki Daishido for identifying the references to

Tanaka's work and selecting and translating these excerpts from Tanaka (1979).

completely different values and meanings. However, had someone like Gamow or Dicke notified us of the significance of our measurements, it would not have taken us 14 years [to detect the CMBR]. This is a bit of regret.

Medd and Covington (1958) reported discussions with Tanaka, and an improved measurement, 5.5 K, with a probable error of about 6K, at 10.7-cm wavelength. This is close, but it is also consistent with no background microwave radiation.

In the Soviet Union, Shmaonov (1957) reported a study of emission properties of a dish reflector antenna and receiver at 3.2-cm wavelength. Rapid switching to a reference load gave the instrument excellent stability. Our impression is that he could have detected the CMBR if he had thought to do it, though that would have required closer attention to the suppression of radiation from the ground and accounting for radiation from the atmosphere.

In France, measurements that placed a bound on the background radiation temperature at 33-cm wavelength are recalled by one of the authors, James Lequeux, who writes

In the winter of 1954-55, we measured the paraboloid dish antenna pattern of a former German "Wiirzburg" radar equipped with a 33-cm receiver built by Le Roux, that we used for mapping the Galaxy. This involved measuring the signal received from a remote transmitter while pointing the antenna in various directions. Then we calculated the contribution of the ground and the atmosphere to the antenna temperature as a function of the direction pointed by the antenna, and compared to observation (far from the galactic plane, of course). The observed antenna temperature was calibrated with blackbodies. Then we concluded that any contribution from the sky would be less than 3 K, and would be rather uniform. This is what is published in the Comptes Rendus; and signed by

Le Roux.

Given our equipment, and in spite of careful measurements, it would have been foolish to claim a positive detection. Our remote antenna lobes were considerably stronger than those of the horn used by Penzias and Wilson. Thus in the Comptes Rendus paper we only claim an upper limit for the CMBR, admittedly close to the actual value, but only an upper limit.

In contemporary reports of these measurements, Denisse, Lequeux and Le Roux (1957) estimate that T0 is less than about 3K, while Delannoy et al. (1957) conclude (in a translation and commentary kindly provided by Lequeux) "Delannoy et al. write on page 236 of their paper, 'We may only conclude that the temperature of the sky at 900 MHz is certainly not larger than about 20 degrees Kelvin.' And the footnote says 'A stricter upper limit that was proposed [in Le Roux's thesis, 1956] underestimated the errors on the measurements of the antenna beam.'" We emphasize these comments because Le Floch and Bretenaker (1991) have suggested that this experiment may have yielded a detection rather than a limit. The evidence we have seen argues against it.

Jasper Wall (p. 280) describes an experiment in the 1960s that may have been capable of detecting the CMBR, if they had thought to do it. At the relatively long wavelengths of their observations, the radiation from our Galaxy is large and would have complicated the interpretation. But at the time of writing the art of measuring the CMBR energy spectrum at long wavelengths is still under development.

To summarize, our reading of the evidence is that the Bell Laboratories measurements (De Grasse et al. 1959; Ohm 1961) we discussed starting on page 49 were the first likely - though at the time not recognized - direct detection of the CMBR to be presented in the literature.

Doroshkevich and Novikov (1964), whose recollections begin on page 99, likely were the first to understand the importance of the early Bell Laboratories measurements for cosmology. This grew out of their study of the amount of electromagnetic radiation that would be expected to have accumulated from all known sources of radiation. They compared this to what was known about the brightness of the sky after elimination of radiation from local sources on Earth, in the Solar System, and in the Milky Way Galaxy.44 That led them to make three important points. First, the cosmic radiation from known sources - starlight and radio-luminous galaxies - is minimum at wavelengths near 1 mm to 1 cm. Second, measurements near this minimum, at microwave wavelengths, "are extremely important for experimental checking of the Gamow theory," because the fossil radiation would peak up at these wavelengths.45 Third, there already is a useful microwave measurement, from Ohm (1961).

44 Others had been considering this. Shakeshaft (1954) had compared the measured mean brightness of the sky at radio wavelengths, A ~ 3 m, to observed counts and radio luminosities of galaxies. He concluded that the mean radio sky brightness could be produced by the galaxies if, as was becoming clear, some are intense sources of radio radiation. Estimates of the mean energy density in intergalactic starlight, taking account of the shift to the infrared and the loss of energy by the cosmological redshift, were presented in increasing detail in Bondi (1952), McVittie and Wyatt (1959), and Sandage and Tammann (1964). Comments on the considerable challenge of measuring this cosmic infrared background radiation are, in increasing detail, in Baum (1956), Roach (1964), and Harwit (1964). Harwit's analysis of the problem of separating extragalactic starlight from the zodiacal light (sunlight scattered by, and absorbed and reemitted by, interplanetary dust) is recalled on page 329. Progress in the increasingly focused work of checking consistency of the measured sky brightness as a function of wavelength with what is known and conjectured about sources of radiation and their evolution is reviewed in Hauser and Dwek (2001).

45 When the microwave background radiation was identified, at about one percent of the wavelength Shakeshaft was considering, the question he addressed naturally arose again: could this microwave radiation have come from sources in the universe as it is now? The coincidence Doroshkevich and Novikov noted, that the foreground radiation from known sources in galaxies is minimum at microwave wavelengths, aided recognition that the microwave background is

Doroshkevich and Novikov's reference for the "Gamow theory" is Gamow (1949). As it happens, that is the same paper Osterbrock and Rogerson (1961) cited in their comment about the "explosive formation picture" quoted on page 59. But Osterbrock and Rogerson discussed helium, while Doroshkevich and Novikov discussed radiation. Doroshkevich and Novikov were members of Zel'dovich's research group in Moscow. We noted (p. 35) that Zel'dovich (1963a) recognized the importance for cosmology both of helium and the microwave background radiation, but that his reading of the evidence at the time led him to conclude that the big bang likely was cold, not hot.

The recollections in the next chapter make it clear that some astronomers in the early 1960s remembered the evidence we have reviewed (p. 44) that the spin temperature of the interstellar molecule CN is surprisingly large, and that that suggested the presence of a microwave radiation background at a temperature of a few degrees above absolute zero. But the connection to the hot big bang picture seems to have been made only after the radiation had been recognized in direct detection.

Telescopes Mastery

Telescopes Mastery

Through this ebook, you are going to learn what you will need to know all about the telescopes that can provide a fun and rewarding hobby for you and your family!

Get My Free Ebook

Post a comment